This Article Abstract Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (6) Citing Articles via Google Scholar Google Scholar Articles by Blumberg, L. J. Articles by Nichols, L. M. Search for Related Content PubMed PubMed Citation Articles by Blumberg, L. J. Articles by Nichols, L. M. Related Collections Insurance - Employer-Based System

Health Tracking

MARKETWATCH

The Health Status Of Workers Who Decline Employer-Sponsored Insurance

Abstract

 
This paper uses data from the 1997 National Health Interview Survey to compare workers who decline employers’offers of health insurance (decliners) with comparison groups of workers who take up offers of employer coverage and those who do not have such offers. Uninsured decliners fare much worse than coverage takers on every mental health measure. While the evidence on physical health measures is somewhat mixed, decliners who are not healthy appear to have greater difficulty obtaining needed services than do workers who take up employer coverage, although decliners tend to have somewhat better access than do the uninsured who are not offered such coverage.

A number of researchers have studied whether the decline in the rate of employer coverage is the consequence of changes in the rate at which firms offer insurance to their workers, the rate at which workers take up offered coverage, or both. While there are clearly differences across subgroups of workers, recent research demonstrates that there has been a clear decline in the take-up rate among all workers over the past ten years.

The objective of this analysis is to provide some insight into the characteristics of "decliners." Our primary focus is on workers who decline offers of employer coverage and are uninsured as a result, since they are of interest from a policy perspective. We develop a profile of decliners in terms of their socioeconomic characteristics and health status. We also compare them with workers who are not offered health insurance and workers who are offered coverage and take it up.⁴

Data source. This analysis uses data from the 1997 National Health Interview Survey (NHIS). The NHIS, conducted by the National Center for Health Statistics, is a cross-sectional, in-person interview survey of 39,832 households. It is nationally representative and has been conducted annually, in some form, since 1957. It focuses on health status, use of health services, and morbidity; it has a large enough sample to be able to provide reliable estimates of many health conditions by considerable socioeconomic detail. These rich data on health status are unique for a data set that also includes information on employer-based insurance offers and on coverage.

Socioeconomic Profile

Top Socioeconomic Profile Health Status Profile Policy Implications Future Research Needs NOTES

 
To develop a clearer understanding of the nature of the problems facing decliners, we compared their characteristics with those of workers in two comparison groups: workers who are offered coverage by their own employers and who take up that coverage ("takers"); and workers who do not receive a coverage offer from their own employers ("not offered").

When we examined the distribution of health insurance status for workers in each of the three groups (our study group of decliners, plus the two comparison groups), we found that the vast majority of decliners either had employer coverage through a source other than their own employer (63 percent) or were uninsured (25 percent). At 4 percent, the only other insurance status category that represented more than 2 percent of decliners was "private, nongroup."⁵ Of workers who were not offered employer coverage, only 40 percent had employer coverage through another source, while 41 percent were uninsured. Eleven percent of this population had private, nongroup coverage, and about 4 percent had Medicaid (no other category was above 2 percent).

Exhibit 1 shows the time since last insured for currently uninsured workers. Uninsured decliners tended to have had some coverage more recently than the uninsured not offered coverage. About twice as many not-offered workers relative to decliners reported that they have never had health insurance. Uninsured decliners also were more than three times as likely as the uninsured not-offered to report that it had been less than six months since they last had coverage. A majority of both groups, however, had been without coverage for a year or more.

As one would expect, uninsured decliners were more likely than takers, but less likely than the not-offered, to be employed by small firms (fewer than fifty workers). Decliners also fell between the other two groups in terms of percentage employed in the biggest firms (1,000 or more workers).

Health Status Profile

Top Socioeconomic Profile Health Status Profile Policy Implications Future Research Needs NOTES

 
Uninsured decliners reported having worse health status than did takers, and these results are consistent when comparing workers of given ages within each group (Exhibit 3). Decliners also were more likely than takers, but less likely than the not-offered, to have a family member in fair or poor health. These findings seem counterintuitive since one would expect those with health problems to place a greater value on coverage and therefore to be more likely to take it up when offered. The correlation between health status and income, and the consequently reduced purchasing power of the ill compared with the healthy, may explain this result.

Decliners were in worse health than takers, yet they went longer periods without seeing a health professional. Almost 29 percent had gone one year or more without such contact, compared with about 18 percent of takers.

We now turn to a more detailed analysis of health status, using the health conditions information available in the NHIS. We first determined which conditions included in the NHIS affected at least 2 percent of the general population. We then used this limited list of conditions to assess the relative health status of decliners.

Mental health conditions. In addition to physical health conditions, the 1997 survey included questions related to mental health conditions and access to medical services (Exhibit 4). The most startling differences are in the incidence of mental health conditions. Uninsured decliners and the not-offered suffered from all measured mental health problems at a significantly higher rate than was true of coverage takers. The uninsured groups also were more likely to report that their condition(s) interfered with their life.

Physical conditions. Uninsured decliners were more likely than coverage takers were to report having ever had asthma, recent face pain, and migraine headaches. They were, however, less likely to report ever having had diabetes or recent joint pain. Because of the high expected use of medical services for persons with diabetes, those offered coverage surely were likely to take it, which explains the differences between groups for that condition.

Health behavior and general measures of access. The 1997 NHIS asked respondents if their health status was better than, the same as, or worse than it was one year ago. There were no significant differences among the three groups in the share reporting that their health had improved; however, takers were less likely than uninsured decliners were to report being worse off.

Another indicator of health status is the extent to which people make efforts to reduce unhealthy behavior. In that regard, we measured the likelihood that persons who had ever smoked (at least 100 cigarettes over their lifetime) were still smoking. Uninsured decliners who ever smoked were significantly more likely to still be smoking than were their counterparts in the taker group.

Decliners also were more likely than takers were to have had difficulty with access to medical services, although they did fare better on certain measures than did those not offered coverage. Decliners were less likely than takers were to have a usual source of care, but more likely to have one than those not offered coverage were. Decliners were more likely than takers were to be unable to afford drugs, mental health services, and dental care in the past year and to have had at least one emergency room visit. They were less likely than takers were to have had a flu shot.

These results indicate that uninsured decliners who are not healthy have difficulty obtaining needed services. They also have at least a somewhat greater tendency toward unhealthy behavior than those taking up employer coverage and may be less likely to seek out preventive care.

Policy Implications

Top Socioeconomic Profile Health Status Profile Policy Implications Future Research Needs NOTES

 
These results are somewhat surprising, for we expected workers voluntarily declining coverage to be healthier than those enrolling. The only clinical conditions that were significantly more prevalent among employer coverage enrollees were diabetes and chronic joint pain. Since other conditions, especially less than very good overall health status and those reflecting substandard mental health, appeared at a higher rate among decliners, we infer that workers perceive the implications of those maladies quite differently than they view diabetes and joint pain.

Yet we know from the results on bed days and work-loss days that these other conditions are indeed debilitating to uninsured decliners. Our interpretation is that even though they are on average in worse health than workers who take their employer’s offer, the presence of most conditions does not change the fundamental calculus: To these workers, the marginal value of insurance is lower than the out-of-pocket cost they would face if they were to become insured.

This judgment has many possible implications, some of which can only be explored in future multivariate work. However, these descriptive results support the inference that despite reduced access vis-à-vis takers, most decliners are able to obtain what they consider to be minimally adequate care. Almost one-half went to the doctor in the past year, and more than 80 percent reported that they could afford the prescription drugs they needed in the past year. While 36 percent (not shown) had some kind of mental health problem, only 4.5 percent reported that they could not afford mental health care (although some reporting mental health problems may not have sought mental health services). Since they were younger, less educated, and more likely to have low incomes than those who took employer coverage in 1997, it is possible that they had lower demand for both health insurance and health care (or higher thresholds of discomfort) than those who took their employers’ coverage offers. It is also likely, of course, that subsidies would induce some of them to buy health insurance, but perhaps not all of them, voluntarily, ever.

The main policy implication that we draw is that workers without offers should get the highest priority in government efforts to extend coverage. Their health status is practically identical to that of uninsured decliners—save for hypertension and asthma, where they are healthier—and those without offers are even less likely to have seen a doctor lately, to have a usual source of care, and to have engaged in such elementary preventive behavior as getting a flu shot. And those without offers account for the bulk of the working uninsured.

This is not to say that decliners deserve no subsidy. A strong case can be made that all low-income persons should get new subsidies to purchase health insurance, on equity and efficiency grounds.⁷ Decliners with health problems must feel somewhat income constrained. Also, we believe that any new subsidies, such as individual tax credits, should be applicable to employee premiums as well as to products in the nongroup market.⁸ This is the best way to ensure that firms that offer coverage now continue to do so, and thereby to preserve today’s limited degree of risk pooling in private health insurance markets.⁹

The importance of mental health conditions to the relatively poor health status of uninsured decliners, despite the small percentage who reported difficulty paying for mental health care, suggests two policy responses. First, outpatient mental health services may need to be given a higher priority than they have traditionally been accorded by safety-net providers. Second, monitoring and maintaining mental health parity in group insurance may be more important to employee take-up than was previously believed. Mental health coverage is not cheap, but our results suggest that a substantial fraction of uninsured decliners with mental health problems may not even be seeking care.

Of course, bringing these persons into the insured risk pool may raise costs overall, but spread over the entire insured population, the extra cost entailed per insured person should be fairly small.¹⁰ Mental health parity may partially or even completely pay for itself with fewer work days lost and higher productivity overall. The recent proliferation of employee assistance programs—employer-financed systems intended to help employees through particularly stressful personal life changes—would attest to employers’ perceptions of the productivity lost to the temporary mental health needs of workers.

Future Research Needs

Top Socioeconomic Profile Health Status Profile Policy Implications Future Research Needs NOTES

 
Policymakers clearly need researchers to deliver a more nuanced understanding of why so many workers in less-than-perfect health decline their employer’s offer of health insurance. Future multivariate work should devote itself to this task. Cost clearly matters, but the particular willingness to pay by workers with health problems is of great interest. A subsidy large enough to bring employee premiums to a level at which these relatively needy workers would purchase coverage would likely be a very efficient incremental coverage strategy.

At the same time, it would be ideal to study the longitudinal health status of those who decline coverage in any given year and are in less-than-excellent health. We note that uninsured decliners were more likely to have had deteriorations in their health status in the past year than takers were (Exhibit 4). At the same time, almost 60 percent of uninsured decliners had been without health insurance for more than a year (Exhibit 1). What are the health consequences of being uninsured for a long time, and are decliners aware of them? The answers to these questions will have important ramifications for policy and for efforts to educate workers about the value of health insurance coverage.

Editor's Notes

 
Linda Blumberg is a senior research associate at the Urban Institute, in Washington, D.C. When this was written, Len Nichols was a principal research associate there. He is now the vice-president of the Center for Studying Health System Change, in Washington, D.C.

This research was funded by a grant from the U.S. Department of Health and Human Services (DHHS), Assistant Secretary for Planning and Evaluation (ASPE)/Health Policy. The authors thank Holly Harvey and Bernadette Fernandez for their helpful comments, Stacey McMorrow for research assistance, Singha Sarbajit for programming, and two anonymous reviewers. The opinions expressed herein are those of the authorsdo not necessarily reflect those of the DHHS, the Urban Institute, or its funders.

NOTES

Top Socioeconomic Profile Health Status Profile Policy Implications Future Research Needs NOTES

 

1. D.W. Liska, N.J. Brennan, and B.K. Bruen, State-Level Databook onHealth CareAccess and Financing, 3d ed. (Washington: Urban Institute, 1998).
2. P. Fronstin, Employment-Based Health Benefits: Trends and Outlook, EBRI Issue Brief no. 233 (Washington: Employee Benefit Research Institute, 2001).
3. P.F. Cooper and B.S. Schone, "More Offers, Fewer Takers for Employment-Based Health Insurance: 1987 and 1996," Health Affairs (Nov/Dec 1997): 142–149; and H.S. Farber and H. Levy, "Recent Trends in Employer-Sponsored Health Insurance Coverage: Are Bad Jobs Getting Worse?" Journal of Health Economics (January 2000): 93–119.
4. This paper is based upon a longer report funded by the U.S. Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation, Decliners of Employer-Based Health Insurance: An Analysis Using the National Health Interview Survey (Washington: Urban Institute, October 2000).
5. Insurance status categories included the following: Medicare and employer coverage; coverage from own employer; employer coverage from other source; employer coverage of unknown type; Medicare, no employer coverage; private, nongroup coverage; private, type unknown; Medicaid; other public; Civilian Health and Medical Program of the Uniformed Services (CHAMPUS); military; Indian Health Service; and uninsured.
6. The remaining analyses exclude decliners and those not offered employer coverage who obtain coverage through another source. We chose to focus on the uninsured, because their situations are of the greatest policy interest. Those who take up employer coverage offers are an appropriate comparison group since they work for employers that offer coverage, just as decliners do, and face the same take-up decision as the decliners but have a different outcome. In this way, the interpretation of comparisons of workers in different groups is not complicated by the fact that some have access to employer coverage through their spouse.
7. L.J. Blumberg and L.M. Nichols, "Why Are So Many Americans Uninsured?" (Paper presented at the University of Michigan Research Agenda Setting Conference, July 2001).
8. See J. Gruber and L. Levitt, "Tax Subsidies for Health Insurance: Costs and Benefits," Health Affairs (Jan/Feb 2000): 72–85.
9. L.J. Blumberg and L.M. Nichols, Decisions to Buy Private Health Insurance (Report to the U.S. Department of Labor, Pension and Welfare Benefits Administration, 2000).
10. Congressional Budget Office. "Cost Estimate, S. 543, Mental Health Equitable Treatment Act of 2001," 22 August 2001, <www.cbo.gov/showdoc.cfm?index=3013&sequence=0&from=6>(28 September 2001).

                      What's this?

This article has been cited by other articles:

				M. A. Hall The Death of Managed Care: A Regulatory Autopsy Journal of Health Politics Policy and Law, June 1, 2005; 30(3): 427 - 452. [Abstract] [PDF]

				T. C. Buchmueller, K. Grumbach, R. Kronick, and J. G. Kahn Book Review: The Effect of Health Insurance on Medical Care Utilization and Implications for Insurance Expansion: A Review of the Literature Med Care Res Rev, February 1, 2005; 62(1): 3 - 30. [Abstract] [PDF]

				G. A. Jensen and M. A. Morrisey Are Healthier Older Adults Choosing Managed Care? Gerontologist, February 1, 2004; 44(1): 85 - 94. [Abstract] [Full Text] [PDF]

				R. D. Galloway Health Promotion: Causes, Beliefs and Measurements Clin. Med. Res., July 1, 2003; 1(3): 249 - 258. [Abstract] [Full Text] [PDF]

				J. C. Robinson The Politics of Managed Competition: Public Abuse of the Private Interest Journal of Health Politics Policy and Law, April 1, 2003; 28(2-3): 341 - 354. [Abstract] [PDF]